Wen Haihu, from Nanjing University, said he was “totally shocked by the results and the data” after watching Dias’ talk online and immediately asked his team to try and replicate the study, he told China Science Daily.

After working nonstop for eight days and nights, Wen and his colleagues were unable to achieve superconductivity at the US team’s reported 21 degrees Celsius (69.8 Fahrenheit) and 1 gigapascal of pressure.

The first hurdle came with the team’s attempt to make samples of the material – a combination of the rare earth element lutetium, along with hydrogen and a little bit of nitrogen – by following the protocol described in the US paper.

“They might have left out a zero or something in the preparation temperature, which is obviously too low for such a synthesis reaction to happen,” Wen said.

The researchers came up with their own preparation method and made sure the colour, structure, and composition of the samples were consistent with the US team’s report.

After measuring the samples’ key properties – such as electrical resistance and magnetisation – under a wide range of pressures and temperatures, “our study showed no evidence of superconductivity in this type of nitrogen-doped lutetium hydride under room temperature and near-room pressure”, Wen said.

In another attempt, a group from the Institute of Physics, Chinese Academy of Sciences in Beijing made a lutetium hydride which only showed superconductivity at -203 degrees Celsius (-333.4 Fahrenheit).

01:20

China sets new world record in development of ‘artificial sun’

China sets new world record in development of ‘artificial sun’

A separate group from the same institute said it reproduced the colour changes reported by the US team, with its samples turning from dark blue to pink and red as pressures rose. However, no superconductivity was observed, down to -271 degrees Celsius (-455.8 Fahrenheit).

Physicist Jorge Hirsch from the University of California San Diego said he was not surprised by these results. “This casts further strong doubt on the validity of the results published in Nature,” he said.

A spokesperson from Nature said they were “encouraged” that the community moved so quickly in attempts to replicate the findings, and that they would “follow these discussions closely”.

“We view attempts at replication of studies published in Nature as an extremely important part of the scientific process. An inherent principle of publication is that others should be able to replicate and build upon the authors’ published claims.”

The hunt for materials that lose their electrical resistance when they are cooled to near absolute zero – a state known as superconductivity – has been going on for years and could revolutionise the efficiency of power grids, computer chips, high-speed trains and medical imaging.

While a few superconductors such as hydrogen sulphide can work under warmer conditions, they must be squeezed to extremely high pressures – beyond a million atmospheres, or 100 gigapascals.

Wei’s team measured their samples’ electrical resistance at different pressures below 6 gigapascals. They found no trace of superconductivity even as the temperature dropped to -263 degrees Celsius, according to their manuscript which has not yet been peer-reviewed.

They also tried to detect signals related to the unique diamagnetic behaviour of superconductors, which expel magnetic fields during their transition to superconductivity, called the Meissner effect.

No diamagnetic signal was detected within the claimed pressure and temperature ranges, the researchers said.

But the Nanjing University study implied a possible alternative cause for the phenomena ascribed to superconductivity by the US team, said Dirk van de Marel, an honorary professor of physics at the University of Geneva in Switzerland.

“It’s in part modifications of the raw data, and in part incorrect interpretation of a phenomenon known as hydrogen ordering that has nothing to do with superconductivity,” he said.

Van de Marel said he received a preprint of the US paper from about 10 journalists simultaneously, seeking his comments in the week before publication, and his first impression was favourable.

“The Dias team demonstrated superconductivity from resistivity, susceptibility, Meissner effect and specific heat. To have all that in a single paper constitutes an extraordinarily well-documented report on superconductivity.”

However, as more scientists started to take a closer look at the paper’s data, “the downfall started”, van de Marel said. He believes it is possible that most of the US team’s data provides “the illusion of a superconducting phase transition”.

While more research is needed to precisely pinpoint the properties of the materials, van de Marel said the samples made by Wen’s team may have contained a higher concentration of nitrogen than the US study.

“In principle, it’s possible that a very precise dose of nitrogen is needed to make these materials superconducting,” van de Marel said.

In the interview with China Science Daily, Wen said his team had started making additional measurements to see if there could be superconductivity under higher pressures, such as tens of gigapascals.

“We hope more researchers from around the world will join us to do the replication work,” he said.

Hirsch said further work by the Chinese teams and others may conclusively establish that the claims of room-temperature, near-room pressure superconductivity are invalid.

But van de Marel was less sure. “Provided an experimental result is true, the process of confirming it by other experiments can be a quick process with a bit of luck. However, it can take forever to show that a claimed result does not exist,” he said.

Both Hirsch and van de Marel have called for Dias to share his samples with other researchers to independently verify whether they are superconducting.

Dias has so far not responded to a request for comment but has told several scientific publications, including Science and Quanta, that he has no plans to distribute the material because of intellectual property issues.

Physicist Mikhail Eremets said his group at the Max Planck Institute for Chemistry in Mainz, Germany had not yet tried to replicate the Nature paper’s results, after a “wasted half a year” testing the Dias team’s previous findings.

Eremets said it had taken about two years to do extensive experimental and theoretical work, only to conclude that the claimed superconductivity in the US team’s 2020 paper was likely to be wrong.

“Naturally, we are not enthusiastic to repeat this disappointing experience,” he said.